The present invention discloses a method for producing a pentitol. The present invention relates to a method for producing pentitols from hexoses. Specifically, xylitol is obtained from glucose in a reaction comprising a sequence of only two separate steps. Glucose is fermented to arabinitol and chemically isomerised to a pentitol mixture comprising xylitol.
Xylitol is produced on an industrial scale by hydrogenation of xylose. Xylose is not available as such it is obtained from xylan-containing plant materials. In order to obtain xylose, xylan-containing plant materials such as almond shells, corn cobs or birch wood are hydrolysed in acidic medium at elevated temperatures. This hydrolysis suffers from two major disadvantages: a high load of waste material due to the low content of xylan in the above mentioned starting materials and a low product purity and yield due to considerable formation of by-products under the extreme hydrolysis conditions which are used. Extensive purification and refining is required to remove excess of acid and the pronounced colour. The subsequent crystallisation of the demineralised xylose syrup suffers from the low purity of the xylose syrup. Other hemicellulosic sugars also formed during hydrolysis have similar physico-chemical properties and have to be removed quantitatively. In an earlier stage to avoid the formation of galactitol, galactose has to be removed prior to the catalytic hydrogenation. Application of xylitol in food and related products requires the complete removal of galactitol for reasons of human safety, e.g. eye damage.
For every kilogram of crystalline xylitol 12 to 13 kg of almond shells have to be processed, resulting in about 11 to 12 kg of solid waste. Apart from a pollution problem there is also a logistic problem with this process in that large quantities of almond shells have to be transported. Finally, the availability of the xylan containing material may become a limiting factor.
It is therefore of interest to consider alternative processes for producing xylitol which do not suffer from the mentioned drawbacks. Chemical and microbial processes for producing xylitol have been described.
Recently, some reaction schemes to produce xylitol, starting from readily available hexoses, in particular D-glucose and D-galactose, have been published. All of these schemes comprise a sequence of more than two reaction steps. In a first step the hexose is submitted to a chain shortening reaction which yields a C5-intermediate. This step is performed either fermentatively or chemically. The subsequent process relates to the conversion of the C5-intermediate into xylitol, by using a sequence of at least two fermentative and/or chemical conversion steps.
In EP 403 392 and EP 421 882 a four step process is disclosed in which glucose is fermented to D-arabinitol by an osmophilic yeast. Subsequently, the arabinitol (C5- intermediate) is converted by bacteria (Acetobacter, Gluconobacter or Klebsiella) into D-xylulose. In the third step xylulose is isomerised by glucose (xylose) isomerase into a xylose/xylulose mixture. In the final step either the xylose is enriched prior to hydrogenation by chromatography, or the xylulose/xylose mixture is directly subjected to hydrogenation followed by the separation of xylitol by chromatography.
In WO 93/19030 glucose, fructose or galactose or mixtures thereof (obtained by hydrolysis of the disaccharides sucrose and lactose) are oxidatively decarboxylated into alkali metal arabinonate and lyxonate, respectively. These intermediates are first converted to the aldonic acid form. Subsequently, the aldonic acids which are the C5- intermediates, are transformed into xylitol. When L-sorbose is used, L-xylonate is obtained via the oxidative decarboxylation and this is converted to the aldonic acid form before being hydrogenated to xylitol. This last pathway seems simple however one has to take into account the reaction steps required to obtain L-sorbose. L-Sorbose is mainly obtained via fermentative oxidation of sorbitol, which in turn is obtained from glucose by catalytic hydrogenation, resulting overall in five reaction steps to obtain the final xylitol.
Other chemical methods for xylitol preparation include elaborated reaction schemes involving the use of protection groups. Due to the lack of economic feasibility these reactions are not further considered here (Helv. Chim. Acta 58, 1975, 311).
Several exclusively microbiological pathways have been published, however, none of them are competitive because of the overall yield which is far too low.
There exists therefore a need for an economically valuable method for producing pentitols, especially xylitol, with a low level of impurities, which is easily refinable, which comprises a short reaction sequence, and starting from readily available hexoses, such as glucose (anhydrous, monohydrate, or high dextrose syrups).
The present invention provides such a method. The present invention relates to a method for producing a pentitol from a hexose characterised in that the method comprises the following steps,
a) fermentation of a hexose to yield a C5-intermediate consisting mainly of a pentitol,
b) isomerisation of the pentitol of step a) in the presence of a chemical catalyst to yield a corresponding pentitol mixture,
c) optionally separation of the desired pentitol from the product of step b).
The present invention can be summarised as follows. The invention discloses the fermentation of C6-carbohydrates which results in C5-polyols, the fermentation step is followed by chemical catalytic isomerisation. The starting material can be any easily available C6-carbohydrate, the preferred substrate is glucose, anhydous as monohydrate or in the form of a high dextrose syrup. Starting with glucose the fermentation yields mainly arabinitol. The fermentation of the present invention is based on methods known in the art. In carrying out a process according to the present invention, any yeast which has an ability to produce D-arabinitol from glucose may be used. For example yeasts belonging to the genera Pichia, Endomycopsis, Hansenula, Debarvomyces, Zygosaccharomyces, Saccharomyces, Candida and other yeasts belonging to the genus Torulopsis are suitable for this particular fermentation.
The yield of D-arabinitol in the fermentation product is preferably larger than 20% (w/w) more preferably larger than 40% (w/w) based on the initial hexose content. In general with the use of osmophylic yeasts D-arabinitol is the only pentitol which is produced.
The D-arabinitol is subjected to catalytic isomerisation by methods known in the art. D-arabinitol is treated at temperatures between 70 and 250xc2x0 C., preferably at a temperature above 100xc2x0 C., and at hydrogen gas pressures between 0.1 and 10 MPa, preferably between 1 and 8 MPa.
The catalytic isomerisation is performed in the presence of catalysts which are known in the art for perfoming hydrogenation/dehydrogenation. Suitable catalysts include ruthenium, copper, palladium, platinum, rhodium, cobalt and nickel based catalysts, or their oxides and mixtures thereof.
The polyol isomerisation is performed at distinctly different pH levels, and the addition of alkali or acid has an influence on the thermodynamic equilibrium of the pentitol mixture. The isomerisation reaction results in a product containing xylitol, ribitol and DL-arabinitol. Xylitol is present in these mixtures in more than 10% preferably in more than 20%. The reaction product further contains some lower alditols, such as tetritols and triitols, adding up to a maximum of 10% preferably only to 5% of the total alditol content.
The isomerisation mixture is optionally subjected to chromatography on cationic resin material yielding purified xylitol with a purity in excess of 95%. Preferably the mixture is first demineralized and subsequently submitted to chromatography. The refining is suitably performed using a strong cation exchange resin e.g. Duolite C 26 followed by a medium base anion exchange resin Duolite A 368. This process is preferably repeated once. On industrial scale chromatography is performed using suitable equipment obtainable for example from Mitsubishi with Diaion UBK-555 resin (in Ca2+ form).
The other pentitols are optionally recycled to the polyol isomerisation step, resulting in increased overall yield. The xylitol can also be further purified by crystallisation.
The advantage of this process in comparison with earlier described processes such as the process disclosed in WO 93/19030 is that well established unit operations can be used for the refining (classical syrup refining) and that known techniques and equipment for fermentation and catalytic isomerisation can be used. The main advantage compared with other methods such as described in EP 403 392 and EP 421 882 and WO 93/19030 is the much shorter overall reaction sequence. Schematically the method of the present invention is illustrated in Scheme 1, wherein the underlined steps are essential. 